Presbyacusis is a general term to describe hearing loss in older persons or the hearing loss associated with aging. Hearing loss observed in older adults is a result of the combined effects of aging, long-term exposure to occupational and nonoccupational noise, the use of ototoxic drugs, diet, disease, and other factors. In this case, the term presbyacusis describes any hearing loss observed in an older person, regardless of cause. Alternatively, presbyacusis may refer specifically to the hearing loss that increases with chronological age and is related only to age-related deterioration in the auditory periphery and central nervous system (CNS).

Currently, about 75% of the 28 million hearing-impaired individuals in the United States are 55 years of age or older; the number of hearing-impaired individuals will increase as the population ages. Indeed, presbyacusis is the most prevalent of the chronic conditions of aging among men age 65 years and older, and the fifth most prevalent condition among older women, following arthritis, cardiovascular diseases, and visual impairments (National Center for Health Statistics, 1986). Age-related hearing loss in the United States has been well characterized by epidemiologic surveys such as the Framingham Heart Study (Moscicki et al., 1985), the Baltimore Longitudinal Study of Aging (Pearson et al., 1995), and the Epidemiology of Hearing Loss Study of the adult residents of Beaver Dam, Wisconsin (Cruickshanks et al., 1998). Figure 1 shows the systematic increase in thresholds with chronological age in Framingham subjects; thresholds at high frequencies are higher for male than for female subjects. In the Beaver Dam study, the prevalence of hearing loss was 45.9%, with hearing loss defined as average thresholds (0.5–4 kHz) greater than 25 dB in the worse ear. This is consistent with the prevalence reported in the Framingham study of 42%–47%. The prevalence of hearing loss in the Beaver Dam study varied greatly with sex and age, ranging from 10.2% for women 48–52 years of age to 96.6% for men 80–92 years of age. Another set of data, including hearing levels as a function of age, sex, and history of occupational noise exposure, is part of an international standard (ISO 1999, 1990). Database A from this standard may more closely represent aging effects on hearing, given that subjects were screened for occupational and other noise exposure history. Epidemiologic surveys also provide estimates of the genetic component of presbyacusis. Heritability coefficients suggest that as much as 55% of the variance in thresholds in older persons is genetically determined, are stronger in women than in men, and are comparable to those for hypertension and hyperlipidemia (Gates, Couropmitree, and Myers, 1999).

Figure 1..  

Mean pure-tone thresholds (in dB HL) for the better ear of male and female participants in the Framingham Heart Study. Age ranges of participants are given at the right. (Adapted with permission from Moscicki, E. K., et al., 1985, “Hearing loss in the elderly: An epidemiologic study of the Framingham Heart Study cohort.” Ear and Hearing, 6, 184– 190.)

A remarkable and consistent age-related change in hearing occurs at frequencies above 8 kHz and begins as early as age 20–30 years (Stelmachowicz et al., 1989). Figure 2 shows thresholds in the conventional and extended high frequencies for younger and older adults, grouped according to average thresholds from 1 to 4 kHz (Matthews et al., 1997). Thresholds are substantially elevated at frequencies above 8 kHz, even for individuals with nearly normal thresholds at lower frequencies.

Figure 2..  

Mean pure-tone thresholds at frequencies from 0.25 kHz to 18 kHz for three groups of subjects aged 60–79, grouped by pure-tone average at 1, 2, and 4 kHz (normal, mild to moderate, and severe), and one group of younger subjects with normal hearing. (Adapted with permission from Matthews, L. J., et al., 1997, “Extended high-frequency thresholds in older adults.” Journal of Speech, Language, and Hearing Research, 40, 208–214. © American Speech-Language-Hearing Association.)

To supplement pure-tone thresholds, the Hearing Handicap for the Elderly instrument (Ventry and Weinstein, 1982), a self-report questionnaire, has been used to compare older individuals' assessment of their communication abilities with objective measures of hearing and with threshold-based estimates of hearing handicap, such as that recommended by the American Academy of Otolaryngology–Head and Neck Surgery (AAO, 1979; Matthews et al., 1990). Discrepancies between objective and subjective measures are common, and the variance in pure-tone thresholds within hearing handicap categories is large. Thus, whereas hearing loss is among the most prevalent chronic conditions of aging, the impact of hearing loss on communication abilities and daily activities of older adults varies greatly among individuals and is not accurately predicted from the pure-tone audiogram.

Studies of auditory behavior in older adults must separate age-related effects from those attributable simply to reduced audibility resulting from elevated thresholds. One experimental method to minimize the confound of reduced audibility is to include only older subjects whose pure-tone thresholds are equal to those of younger subjects. When changes in auditory behavior in older adults are observed that are not attributable to reduced audibility, they may be due to age-related changes in the auditory periphery, which provides an impoverished input to a normal auditory CNS, or to the combined effects of an aging periphery and an aging CNS. For many behavioral measures, it may not be possible to differentiate between these outcomes. This is particularly the case for tasks that require comparisons of temporal information across intervals of time or that assess binaural processing.

Other than effects related to their hearing loss, older adults probably do not have increased problems in speech understanding relative to younger individuals, as measured conventionally (i.e., monaurally, under earphones, with highly redundant signals). Indeed, some studies suggest that 70%–95% of the variance in monaural speech recognition scores may be accounted for by the variance in speech audibility (Humes, Christopherson, and Cokely, 1992). Figure 3 shows scores on several speech recognition tests for three age groups with nearly identical (within ∼3 dB) mean thresholds from 0.25 to 8 kHz (Dubno et al., 1997). With hearing loss held constant across age group, speech recognition scores also remained constant. Nevertheless, age-related differences in speech recognition in noise may become apparent in more realistic listening environments, such as in the sound field with spatially separated speech and maskers, with competing sounds that have temporal or spectral dips, or on tasks that require divided or selective attention. It remains unclear if these age-related changes may be attributed to an aging auditory periphery or to the combined effects of an aging periphery and an aging CNS.

Figure 3..  

Mean scores in percent correct ±1 standard error) for six measures of speech recognition. The six measures are word recognition for NU-6 monosyllabic words (PB), maximum word recognition (PB_max), maximum synthetic sentence identification (SSI_max), keyword recognition for high-context and low-context sentences from the SPIN test (SPIN-PH and SPIN-PL), and “percent hearing for speech” (SPIN-HFS). (Adapted with permission from Dubno, J. R., et al., 1997, “Age-related and gender-related changes in monaural speech recognition.” Journal of Speech, Language, and Hearing Research, 40, 444–452. © American Speech-Language-Hearing Association.)

Although older adults are the largest group of hearing aid wearers, their satisfaction with hearing aids is low. More than 75% of individuals who are likely to benefit from a hearing aid do not own one, a gap of approximately 20 million people (LaPlante, Hendershot, and Moss, 1992). Similar results were observed for hearing-impaired participants of the Framingham and Beaver Dam studies, in which less than 20% were hearing aid users. In examining the functional health status and psychosocial well-being of older individuals, Bess et al. (1989) concluded that hearing loss is a primary determinant of function and that its impact is comparable to that of other chronic conditions affecting this population. Thus, untreated hearing loss can have a negative effect on quality of life beyond that due to poorer communication abilities (Mulrow et al., 1990). The potential benefit to communication and quality of life, together with new fitting options and improved technology, suggests that older adults should be encouraged to use amplification.

Evidence of age-related changes in the auditory system is revealed in the physiological properties of aging humans and animals. Older gerbils raised in quiet have elevated thresholds of the compound action potential (CAP) of the auditory nerve and shallow slopes of CAP input-output functions (Hellstrom and Schmiedt, 1990). These characteristics are also reflected in higher auditory brainstem response (ABR) thresholds and shallower slopes of ABR amplitude-intensity functions relative to young gerbils (Boettcher, Mills, and Norton, 1993). Similar findings have been observed in older humans. Although these potentials produced by short-duration signals are reduced in amplitude with age, amplitudes of potentials arising from higher CNS centers in response to long-duration signals, such as steady-state potentials and N₁₀₀–P₂₀₀, may be unaffected or even increase with age. Abnormal recovery from adaptation or forward masking and abnormal gap detection at the level of the brainstem have also been observed in older animals and humans (Walton, Orlando, and Burkard, 1999). In aging gerbils, the 80–90 mV dc resting potential in the scala media of the cochlea, known as the endocochlear potential (EP), is reduced substantially. In contrast to these changes, nonlinear phenomena remain relatively intact. For example, transient otoacoustic emissions, reflecting the functioning of outer hair cells, are present in about 90% of older humans with normal hearing, but amplitudes are reduced in a manner that is not predictable by either age or pure-tone thresholds. In aging gerbils, distortion product otoacoustic emissions are present and robust, but somewhat reduced in amplitude (Boettcher, Gratton, and Schmiedt, 1995). Two-tone rate suppression is observed in older gerbils, with age-related threshold shifts (Schmiedt, Mills, and Adams, 1990), although older humans may have reduced suppression measured psychophysically (Dubno and Ahlstrom, 2001).

The pathologic anatomy underlying these physiologic changes was most extensively described through studies of human temporal bones by Schuknecht (1974). Initially, four categories of presbyacusis were identified, including sensory (degeneration of sensory cells), neural (largely loss of spiral ganglion cells), metabolic (degeneration of the lateral wall and stria vascularis, with reduction in the protein Na,K-ATPase), and mechanical (aging of sound-conducting structures of the inner ear). Categories of presbyacusis were revised by Schuknecht and Gacek (1993) wherein atrophy of the stria vascularis was designated as the “predominant lesion” of the aging ear, neuronal loss was “constant and predictable,” mechanical loss remained theoretical, and sensory presbyacusis was the “least important type of loss.” These histopathological findings are consistent with the physiological evidence described above, such as reduced CAP amplitudes and reduced EP but robust otoacoustic emissions. Thus, most age-related changes in hearing can be accounted for by changes observed in the auditory periphery. Nevertheless, there are many age-related anatomical, neurochemical, and neurophysiological changes in the CNS. One prominent neurochemical change is a loss of gamma-aminobutyric acid, which may affect the balance of inhibitory and excitatory neurotransmission. The effects of these and other CNS changes on age-related hearing loss may be substantial but remain largely unknown.